Amazon S3

Learn how to enable Amazon S3 for supported destinations.

Requirements

You must have access to an Amazon Web Services Identity Access Management principal with permissions to:
- Upload objects to an S3 bucket
- Generate a Secret Key and API Key for a role with read access to an S3 bucket
You may also wish to have permissions to create roles with access to specific buckets or to delegate their access.

Set up Amazon S3 as a data source

Apply a use case now

The following sections walk you through a single procedure to link a data source to a use case. Click on your use case to expand the steps. To link the data source only, and apply a use case later, see Apply a use case later.

Customer Match

Through the “Audience manager” page

Offline conversion import and enhanced conversions for leads

Through the “Conversions” page

Tip: Upgrade from offline conversion import to enhanced conversions for leads for more accurate reporting, easier data import, and engaged-view and cross-device conversion attribution.

To set up enhanced conversions for leads, you can either use an existing offline conversion import conversion action or create a new conversion action (recommended if you are switching to a new upload method like Data Manager).

In Google Ads, click Goals > Summary > + New conversion action.
Click Import.
Select CRMs, files, or other data sources.
Select a tracking method.
Under Data source, select Connect a new data source.
1. If you previously linked a data source from the Data manager screen that you want to use, select Select an existing source.
Search and select Amazon S3 from the list of data sources.
Select Direct connection, then click Continue.
Read and accept the customer data policy, then click Continue.
Select a category from the conversion goal list, then click Set up.
Enter your Amazon S3 secret access key, then click Connect.
Click Next.
Enter the URI to the file you want to upload, relative to the host address, then click Next. See supported file types under Requirements, earlier in this article.
Optional: You can apply filters to define the conversion events that you’ll import. This is important especially if you have multiple different conversion goals in the file you will be importing. Learn more about using filters.
To map fields, select the data source fields from the dropdown lists that match the destination fields.
Optional: Apply transformations to your data.
Click Next.
Enter a name for this connection.
Optional: Edit connection details.
Click Finish.
Click Save and continue.

If you’re using offline conversion imports, we highly recommend upgrading to enhanced conversions for leads using the Google Ads Data Manager. Offline conversion import is a legacy feature, and it doesn’t have the durability features that enhanced conversions for leads offers.

Implementation varies depending on what you're measuring and your capabilities. Follow our guide to upgrading offline conversion imports to determine the method that best fits your needs.

Apply a use case later

In this type of setup, you complete each part of the connection at different times. This can be useful when you are ready to link a data source, but you are not ready to set up your use case–for example, you aren’t ready to create a customer list.

Consider the following scenario:

Step 1: Dana is an engineer who manages data for your company. Dana sets up the data source to be used for activation in Google Ads. The data source is ready to be associated with a use case.
Step 2: Mahan is a media specialist who needs to measure audience activation. Mahan creates a customer list and then associates it to the data source that Dana has previously set up, to use that data for Customer Match.

Step 1. Initiate the connection to the data source

Step 2. Complete the connection by applying a use case

Troubleshoot errors

Setup errors

Ensure that you are entering the correct account identifier URL.
Ensure that the first line of your object contains the headers—for example, email address and first name.
Double-check your access key ID and secret access key. They should be different.
Ensure that the object in S3 has a file extension. Files without a file extension are rejected.
Ensure that your conversion action has successfully connected to the chosen connection. Otherwise, no data will come in.

Scheduled run errors

The most common cause of scheduled run failures is updated permissions in AWS. Check that the role or account that corresponds to the access key has permissions to read the object.
Ensure that the object still exists and that the name hasn’t changed.

Choose a section to give feedback on

Was this helpful?

How can we improve it?